On September 4th, a twenty-three-year-old Guinean student, M, arrived at ABC, a busy Paris hospital, complaining of fever, night sweats, and fatigue. He had recently returned from Conakry, and it was established that he had been in contact with an ill relative now being treated as a confirmed Ebola patient. A call to the Regional Health Agency, which runs public health surveillance, established that M was a suspected Ebola case, triggering a plan carefully prepared beforehand.

At ABC, M was isolated in a room taped off with warnings. Trained paramedics bundled M into a special ambulance and sped off to a designated treatment unit at the sprawling DEF University Hospital. There, the plan was put into place: a crisis unit of hospital managers activated a phone tree, the infectious disease ward was emptied, and M. was brought to his room as gowned staff members drew blood to test for Ebola and other tropical diseases. A standard IV cocktail was started—saline to treat dehydration, augmented with antibiotics and antimalarial drugs to cover for co-infections.

Unbeknownst to the specialized team mobilized by the case of M, two days earlier, a seventeen-year-old girl, J, was brought to the emergency room of another Paris hospital. She had arrived with her mother after a trip to West Africa to visit relatives. Feverish for five days, she was delirious. After landing at Charles-de-Gaulle, her mother brought her home and when a cold shower didn't help, decided to bring her to XYZ University Hospital. A nurse whisked her into a resuscitation bay where she was met by Dr. A, who ordered bloodwork, an IV and an intubation kit. Meanwhile, J had deteriorated. A blood-tinged fluid had begun to foam at J's mouth. It took twenty minutes before she was stable enough to be transferred to intensive care, twenty minutes in which no one had time to think beyond the "ABCs" of critical care (airway, breathing, and circulation) drilled into emergency workers. In the hectic resuscitation, many were splashed with bodily fluids. No one initially gave thought—and many were unaware—that she had just arrived from West Africa. The next day, life support was withdrawn and J died; by that time, tests confirmed that she had died from cerebral malaria.

The first case above was in fact a drill I attended—an elaborate role-play conducted by the crisis team of the Paris hospitals trust (APHP), a sprawling network of almost forty hospitals totalling 25,000 beds that dates from Napoleon's decision in 1801 to centralize control of hospitals in the French capital.

The second case at XYZ University Hospital was tragically real—and probably the more likely route by which an Ebola patient might encounter the French health care system. The case haunted many of the workers involved, not because of the imagined threat of Ebola, but because the death was so tragic and avoidable had J gotten treatment for malaria while still in Africa.

The point here is not to feed anxieties about an Ebola outbreak coming "here" from "there." As many have pointed out, this is an unlikely scenario—health care systems in the North are structurally disposed to contain epidemics even without any special preparation. The basic apparatus of public health—triage, diagnosis, treatment, infection control—is built into the system, so that even if a patient elides early detection, a needlestick injury occurs, or a glove rips, contamination will be contained by levels of redundancy. In West Africa however, Ebola spread unnoticed for three months, the first case being diagnosed in March 2014 in an infant who had died around Christmas the year before. By then it was too late.

Which brings me to a point made forcefully by Guillaume Lachenal: Ebola happened despite, and indeed as a result of, over a decade of pandemic preparedness efforts costing billions. These efforts not only failed, they produced this Ebola epidemic. As Andrew Lakoff pointed out, the billions poured into a national security apparatus in the name of Global Health were devoted to "preparedness," a nebulous construct that highlighted surveillance and simulation as key to readiness for bioterrorism and other epidemic threats. Huge sums of money were spent on vaccines for epidemics that never materialized. Yet there were already clear and unambiguous signs that the key to preparedness would lie in hospitals. All those efforts devoted to pandemic preparedness did not involve investing in health systems at the front line of epidemics: hospitals.

When Ebola struck, health care workers sickened and died in large numbers. Lack of basic infection control equipment—such as gloves and masks—doubtlessly played a role. Front line workers will inevitably come into contact with Ebola patients since the majority of patients in West Africa come to health care centers with fever as their chief complaint, and, in an epidemic setting, it is difficult to screen out potential cases without a systematic triage mechanism. Hospitals are therefore particularly vulnerable, even more so when understaffed and underequipped.

When Ebola hits, a vicious cycle ensues—as health workers fall ill or are quarantined, those left are even more vulnerable. No wonder then that in Liberia the health system has all but collapsed. Health care facilities, already the weak link in the chain, have become the fault line along which Ebola tracks.

So what comes next? I offer the following as a critical but engaged response to the Ebola epidemic.

Having failed to bolster the region’s hospitals, efforts to eradicate the epidemic should now be substantially shaped by the legacy of HIV. Ebola is a very different epidemic from HIV. But central to both efforts is a focus on mobilizing survivors and those affected by the epidemic. EVD survivors are immune and therefore constitute the perfect pool of caregivers to replace those who have not recovered. There will be strong pressure to use survivors not only as goodwill ambassadors for prevention and treatment efforts but also as front line personnel for ongoing diagnosis and treatment of EVD patients.

The contrast is the hegemonic "camp" model of dedicated treatment facilities initiated by Médecins Sans Frontières (MSF) and a handful of NGOs and missionaries, a model likely to be dramatically scaled up as the US Military, Paul Farmer, and Harvard announce plans to intervene in Liberia. Yet the sagacity of putting all of our eggs into the treatment-camp model might be questioned. Might an alternative be considered, such that families are supported to care for ill family members with food, infection control tools, training and basic medicines? This is a view to which I am spontaneously sympathetic. I suspect that, partially because of the importance of survivors to control efforts, the approach that stresses community mobilization and care will become increasingly emphasized in the months to come.